As can be seen, curves converge at a shop flow time of 32.2 time units, which is also the value of the time in system. This is the result of unrestricted workload norm levels, meaning that jobs do not wait in the pre-shop pool of orders, i.e. release is immediate. As could be expected, in these circumstances, all workload control strategies give the same results. Tighter norms lead to lower values of shop flow times and, up to a point, also of time in system, Fig. 2. The smallest value of time in system, 28.3 time units, is achieved for a shop flow times of 21.8 time units, under the balancing strategy.This represents about 12% reduction in time in system and about32% reduction in the shop flow time in relation to the immediate release situation. For values below this minimum shop flow time, time in system tends to increase substantially. This means that waiting time in the shop floor is partially replaced by the waiting time in the pre-shop pool of orders. Thus, since the time in system is the sum of the pool time and the shop floor time, we may conclude that waiting times in the pool increase more than waiting times on the shop floor